Welcome to Masterful: The Training Platform for Computer Vision Models

Intro

Masterful is the training platform for computer vision models.

Masterful is designed with three objectives:

  1. Accuracy, achieved via comprehensive regularization and semi-supervised learning (e.g. learning from raw, unlabeled images).

  2. Developer Productivity, achieved via high-speed metalearning. It’s a waste of developer time to guess-and-check of hyperparameters or do long runs with a black box optimizer.

  3. Speed, or minimizing GPU-hours and wall clock time, achieved via high-speed metalearning to discover ideal optimization hyperparameters.

_images/arch.png

Users define the two inputs to Masterful:

  • Data

  • Model architecture

Masterful processes these inputs through four modules:

  • Regularization, which improves accuracy from the existing information.

  • Semi-supervised learning (SSL), which improves accuracy by learning from the information in raw, unlabeled images.

  • Optimization, which minimizes GPU-hours and wall-clock time of training.

  • Meta-learning, which minimizes developer “guessing and checking” and eliminates black box optimization.

Masterful returns a trained model (e.g. model weights).

Masterful supports most types of classification, detection, and segmentation.

Currently available for Tensorflow2, with PyTorch support coming soon.

Getting Started

Install with pip install --upgrade pip; pip install masterful.

For detailed installation instructions, visit the Installation Tutorial.

Then execute your first training run with Masterful with the Quickstart Tutorial.

Finally, visualize Masterful’s performance with the Frontend Tutorial.

Design

To learn about the concepts within the Masterful CV Platform, visit the technical report preprint.

arXiv_badge

If you use Masterful for academic research, you are encouraged to cite:

@article{wookeyhorikert2022masterful,
title={Masterful: A Training Platform for Computer Vision Models},
author={Wookey, Samuel and Ho, Yaoshiang and Rikert, Tom and Lopez, Juan David Gil and Beancur, Juan Manuel Mu{\~n}oz and Cortes, Santiago and Tawil, Ray and Sabin, Aaron and Lynch, Jack and Harper, Travis and others},
journal={arXiv preprint arXiv:2205.10469},
year={2022}
}

Documentation

The rest of the documentation is organized into the following sections:

Getting Started

Installation and Quickstart Tutorial.

Image Classification, Detection, and Segmentation

Code examples to help you get started with the Masterful AutoML platform.

Semi-Supervised Learning

Code examples exploring SSL using Masterful.

Advanced Topics

Code examples for advanced use cases.

API Reference

Documenting the classes, methods, and functions.